Method and apparatus for obtaining a durable fault log for a microprocessor

Claim 2

In a microprocessor controlled system, a microprocessor controlled fault logging arrangement for logging at least one of hardware and software faults that occur during an operation of the microprocessor controlled system, comprising: a non-volatile memory; a health check arrangement; a microprocessor connected via a communication bus to the non-volatile memory and to said health check arrangement which produces an error output upon detection of a fault that occurs during operation of the microprocessor controlled system, wherein the error output of the health check arrangement is applied from said health check arrangement through said communication bus to a first input of the microprocessor to cause condition data indicating a condition of a plurality of operational entities associated with the microprocessor to be sent from said microprocessor through said communication bus and to be recorded in the non-volatile memory when the health check arrangement generates the error output; and delay means for delaying the error output of the health check arrangement, said delay means having an input to receive the error output of the health check arrangement, and for delaying the error output, an output for coupling a delayed error output of the health check arrangement to a reset input of the microprocessor, said delay means providing a time delay sufficient to permit the condition data of the plurality of operational entities to be recorded in the non-volatile memory before the delayed error output at the output of said delay means activates said reset input of the microprocessor; and wherein said health check arrangement includes: hardware monitoring means for monitoring operation of said microprocessor controlled system and for producing at least one first fault signal in response to detection of at least one corresponding fault condition; and software monitoring means for monitoring operation of the microprocessor controlled system and for producing at least one second fault signal in response to detection of at least one corresponding fault condition; and the microprocessor controlled fault logging arrangement further including an OR logic means for producing the error output in response to each fault signal.

Claim 3

The fault logging arrangement of claim 1, wherein the first input of the microprocessor is a non-maskable interrupt (NMI) input.

Claim 4

The fault logging arrangement of claim 1, wherein the health check arrangement includes: a watchdog timer circuit connected via said communication bus to the microprocessor, and the watchdog timer circuit being adapted to monitor the operation of the microprocessor, wherein the microprocessor routinely runs a lock-up test routine and transmits a watchdog timer reset signal when the lock-up test routine indicates that the microprocessor is not locked-up; and wherein: the watchdog timer reset signal resets the watchdog timer circuit and prevents the watchdog timer circuit from timing out and generating a watchdog lock-up error signal; and when the lock-up test routine fails to transmit the watchdog timer reset signal within a period determined by the watchdog timer circuit, the watchdog timer circuit generates the lock-up error signal which is applied to the OR logic means to produce the error output.

Claim 5

The fault logging arrangement of claim 3, wherein the first input of the microprocessor is a non-maskable interrupt (NMI) input.

Claim 6

The fault logging arrangement of claim 3, wherein the first input of the microprocessor is a non-maskable interrupt input.

Claim 7

In a method of obtaining a non-volatile log of hardware and software faults that occur during an operation of a microprocessor controlled system that includes a microprocessor, a communication bus, a non-volatile memory, a health check arrangement and an OR logic means, the microprocessor being connected via the communication bus to the non-volatile memory and to the health check arrangement, the improvement comprising: monitoring an operation of the microprocessor controlled system by the health check arrangement; producing an error signal each time one of a hardware and a software fault is detected by the health check arrangement; applying said error signal to the OR logic means to select each error signal; applying each selected error signal outputted from the OR logic means to a non-maskable interrupt input of the microprocessor to initiate a fault condition save routine wherein the microprocessor causes conditions of selected parameters in the microprocessor controlled system to be stored in the non-volatile memory; delaying the error signal produced by the health check arrangement while the fault condition save routine is being carried out; then applying a delayed error signal to a reset input of the microprocessor to cause the microprocessor to reset the microprocessor controlled system after completion of the fault condition save routine; and transmitting at least a portion of the contents of an error log in the non-volatile memory to a remote location for analysis.

Claim 8

In a method of obtaining a non-volatile log of hardware and software faults that occur during an operation of a microprocessor controlled system that includes a microprocessor, a communication bus, a non-volatile memory, a health check arrangement and an OR logic means, the microprocessor being connected via the communication bus to the non-volatile memory and to the health check arrangement, the improvement comprising: monitoring an operation of the microprocessor controlled system by the health check arrangement; producing an error signal each time one of a hardware and a software fault is detected by the health check arrangement; applying said error signal to the OR logic means to select each error signal; applying each selected error signal outputted from the OR logic means to a non-maskable interrupt input of the microprocessor to initiate a fault condition save routine wherein the microprocessor causes conditions of selected parameters in the microprocessor controlled system to be stored in the non-volatile memory; delaying the error signal produced by the health check arrangement while the fault condition save routine is being carried out; then applying a delayed error signal to a reset input of the microprocessor to cause the microprocessor to reset the microprocessor controlled system after completion of the fault condition save routine; and transmitting at least a portion of the contents of an error log in the non-volatile memory to a remote location for analysis.

Claim 9

The method of claim 6, further comprising the steps of: routinely running a lock-up test routine in the microprocessor and transmitting a watchdog timer reset signal from a watchdog timer circuit, which is part of the health check arrangement, when the lock-up test routine indicates that the microprocessor is not locked-up; resetting the watchdog timer circuit with the watchdog timer reset signal, for preventing the watchdog timer circuit from timing out and generating a lock-up error signal; and generating, in the watchdog timer circuit when the lock-up test routine fails to transmit the watchdog timer reset signal within a period determined by the watchdog timer circuit, the lock-up error signal which is applied to the OR logic means from the watchdog timer circuit to produce the error output.

Claim 10

In a microprocessor controlled system, a microprocessor controlled fault logging arrangement for logging at least one of hardware and software faults that occur during an operation of the microprocessor controlled system, comprising: a non-volatile memory; a health check arrangement; a microprocessor connected via a communication bus to the non-volatile memory and to said health check arrangement which produces an error output upon detection of a fault that occurs during operation of the microprocessor controlled system, wherein the error output of the health check arrangement is applied from said health check arrangement through said communication bus to a first input of the microprocessor to cause condition data indicating a condition of a plurality of operational entities associated with the microprocessor to be sent from said microprocessor through said communication bus and to be recorded in the non-volatile memory when the health check arrangement generates the error output; and delay means for delaying the error output of the health check arrangement, said delay means having an input to receive the error output of the health check arrangement, and an output for coupling a delayed error output of the health check arrangement to a reset input of the microprocessor, said delay means providing a time delay sufficient to permit the condition data of the plurality of operational entities to be recorded in the non-volatile memory before the delayed error output at the output of said delay means activates said reset input of the microprocessor; and wherein said health check arrangement includes: hardware monitoring means for monitoring operation of said microprocessor controlled system and for producing at least one first fault signal in response to detection of at least one corresponding fault condition; and software monitoring means for monitoring operation of the microprocessor controlled system and for producing at least one second fault signal in response to detection of at least one corresponding fault condition; and the microprocessor controlled fault logging arrangement further including an OR logic means for producing the error output in response to each fault signal; and wherein the microprocessor is programmed to interrogate an error log corresponding to information of said detected fault conditions and to transmit the error log information to a remote location for analysis.

Claim 11

In a microprocessor controlled system, a microprocessor controlled fault logging arrangement for logging at least one of hardware and software faults that occur during an operation of the microprocessor controlled system, comprising: a non-volatile memory; a health check arrangement; a microprocessor connected via a communication bus to the non-volatile memory and to said health check arrangement which produces an error output upon detection of a fault that occurs during operation of the microprocessor controlled system, wherein the error output of the health check arrangement is applied from said health check arrangement through said communication bus to a first input of the microprocessor to cause condition data indicating a condition of a plurality of operational entities associated with the microprocessor to be sent from said microprocessor through said communication bus and to be recorded in the non-volatile memory when the health check arrangement generates the error output; and delay means for delaying the error output of the health check arrangement, said delay means having an input to receive the error output of the health check arrangement, and an output for coupling a delayed error output of the health check arrangement to a reset input of the microprocessor, said delay means providing a time delay sufficient to permit the condition data of the plurality of operational entities to be recorded in the non-volatile memory before the delayed error output at the output of said delay means activates said reset input of the microprocessor; and wherein said health check arrangement includes: hardware monitoring means for monitoring operation of said microprocessor controlled system and for producing at least one first fault signal in response to detection of at least one corresponding fault condition; and software monitoring means for monitoring operation of the microprocessor controlled system and for producing at least one second fault signal in response to detection of at least one corresponding fault condition; and the microprocessor controlled fault logging arrangement further including an OR logic means for producing the error output in response to each fault signal; and wherein the microprocessor is programmed to interrogate an error log corresponding to information of said detected fault conditions and to transmit the error log information to a remote location for analysis.

Claim 12

The fault logging arrangement of claim 8, wherein the first input of the microprocessor is a non-maskable interrupt input.

Claim 13

The fault logging arrangement of claim 8, wherein the health check arrangement includes: a watchdog timer circuit connected via said communication bus to the microprocessor, and the watchdog timer circuit being adapted to monitor the operation of the microprocessor, wherein the microprocessor routinely runs a lock-up test routine and transmits a watchdog timer reset signal when the lock-up test routine indicates that the microprocessor is not locked-up; and wherein: the watchdog timer reset signal resets the watchdog timer circuit and prevents the watchdog timer circuit from timing out and generating a watchdog lock-up error signal; and when the lock-up test routine fails to transmit the watchdog timer reset signal within a period determined by the watchdog timer circuit, the watchdog timer circuit generates the lock-up error signal which is applied to the OR logic means to produce the error output.